Upregulation of CRABP1 in Human Neuroblastoma Cells Overproducing

BMC Medicine BioMed Central

Research article Open Access Upregulation of CRABP1 in human neuroblastoma cells overproducing the Alzheimer-typical Aβ42 reduces their differentiation potential Markus Uhrig1,2, Peter Brechlin3,4, Olaf Jahn4,5, Yuri Knyazev6, Annette Weninger6, Laura Busia1, Kamran Honarnejad1, Markus Otto7 and Tobias Hartmann*1,2

Address: 1Center for Molecular Biology of the University of Heidelberg (ZMBH), D-69120 Heidelberg, Germany, 2Institute for Neurodegeneration and Neurobiology, Neurology, Saarland University, D-66421 Homburg/Saar, Germany, 3Department of Neurodegeneration and Restorative Research, Center for Neurological Medicine, University of Göttingen, D-37073 Göttingen, Germany, 4DFG Research Center for Molecular Physiology of the Brain (CMPB), D-37073 Göttingen, Germany, 5Max Planck Institute for Experimental Medicine, Proteomics, D-37075 Göttingen, Germany, 6German Cancer Research Center (DKFZ), D-69120 Heidelberg, Germany and 7Department of Neurology, D-89075 Ulm, Germany Email: Markus Uhrig - [email protected]; Peter Brechlin - [email protected]; Olaf Jahn - [email protected]; Yuri Knyazev - [email protected]; Annette Weninger - [email protected]; Laura Busia - [email protected]; Kamran Honarnejad - [email protected]; Markus Otto - [email protected]; Tobias Hartmann* - [email protected] * Corresponding author

Published: 16 December 2008 Received: 3 November 2008 Accepted: 16 December 2008 BMC Medicine 2008, 6:38 doi:10.1186/1741-7015-6-38 This article is available from: http://www.biomedcentral.com/1741-7015/6/38 © 2008 Uhrig et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Abstract Background: Alzheimer's disease (AD) is characterized by neurodegeneration and changes in cellular processes, including neurogenesis. Proteolytic processing of the amyloid precursor protein (APP) plays a central role in AD. Owing to varying APP processing, several β-amyloid peptides (Aβ) are generated. In contrast to the form with 40 amino acids (Aβ40), the variant with 42 amino acids (Aβ42) is thought to be the pathogenic form triggering the pathological cascade in AD. While total- Aβ effects have been studied extensively, little is known about specific genome-wide effects triggered by Aβ42 or Aβ40 derived from their direct precursor C99. Methods: A combined transcriptomics/proteomics analysis was performed to measure the effects of intracellularly generated Aβ peptides in human neuroblastoma cells. Data was validated by real- time polymerase chain reaction (real-time PCR) and a functional validation was carried out using RNA interference. Results: Here we studied the transcriptomic and proteomic responses to increased or decreased Aβ42 and Aβ40 levels generated in human neuroblastoma cells. Genome-wide expression profiles (Affymetrix) and proteomic approaches were combined to analyze the cellular response to the changed Aβ42- and Aβ40-levels. The cells responded to this challenge with significant changes in their expression pattern. We identified several dysregulated genes and proteins, but only the cellular retinoic acid binding protein 1 (CRABP1) was up-regulated exclusively in cells expressing an increased Aβ42/Aβ40 ratio. This consequently reduced all-trans retinoic acid (RA)-induced differentiation, validated by CRABP1 knock down, which led to recovery of the cellular response to RA treatment and cellular sprouting under physiological RA concentrations. Importantly, this

Page 1 of 12 (page number not for citation purposes) BMC Medicine 2008, 6:38 http://www.biomedcentral.com/1741-7015/6/38

effect was specific to the AD typical increase in the Aβ42/Aβ40 ratio, whereas a decreased ratio did not result in up-regulation of CRABP1.

Conclusion: We conclude that increasing the Aβ42/Aβ40 ratio up-regulates CRABP1, which in turn reduces the differentiation potential of the human neuroblastoma cell line SH-SY5Y, but increases cell proliferation. This work might contribute to the better understanding of AD neurogenesis, currently a controversial topic.

Background cells to replace neurons lost in the disease [14]. The effect Alzheimer's disease (AD) is a genetically heterogeneous of AD on neurogenesis has recently been reproduced in a disorder because mutations in multiple genes are transgenic mouse model [15] in which APP mutations involved along with non-genetic factors [1]. The risk may lead to increased incorporation of BrdU and expression of be determined by the effects of numerous loci, some of immature neuronal markers in two neuroproliferative which may produce only minor contributions. Amyloid regions: the dentate gyrus and the subventricular zone. As precursor protein (APP), presenilin1, presenilin2 and the neurogenesis is increased in these mice in the absence of apolipoprotein E ε4 allele have been associated with AD neuronal loss, it might be triggered by more subtle disease [2,3]. These genes are assumed to be responsible for manifestations, for example the initial accumulation of approximately 50% of the genetic background of the dis- the Aβ peptide. In transgenic mice, overexpressing famil- ease, suggesting that further susceptibility genes exist. ial AD variants of APP and/or PS1 dramatically dimin- Genetic analyses of kindred with AD have pointed to β- ished survival of newborn neurons 4 weeks after birth [16]. amyloid peptides (Aβ) as the initiating molecules in the This data hints at an increased neurogenesis in AD, but in development of the disease. contrast to this, also point to early detrimental events shortly after the neurons are born. Biochemical work on APP processing revealed that pathogenic mutations alter processing in such a way that more Methods Aβ42 is produced. Genetic and biochemical data together For details, see the Additional file 1. suggested that Aβ42 accumulation was the primary event in the pathogenesis of AD. Aβ42, but not the more abun- Plasmids dant Aβ40, may cause neuronal dysfunction and trigger C99 encoding sequences were cloned into a pCEP4 vector neurodegeneration in vivo [4,5]. APP is cleaved by β-secre- (Invitrogen) resulting in the following constructs: pCEP4- tase within its ectodomain, resulting in the generation of spA4ct-DA-WT, pCEP4-spA4ct-DA-I45F and pCEP4- the C-terminal fragment C99, which is further cleaved by spA4ct-DA-V50F. The plasmid constructs have been the γ-secretase complex. APP processing results in the described previously [6,7]. release of different peptides. To focus on Aβ, we used the standard construct that maintains APP sorting and the rel- Cell line, cell culture and transfections evant processing events [6,7]. The pathological mecha- Human neuroblastoma SH-SY5Y cells [17,18] were cul- nism of how Aβ42 or Aβ40 acts is unclear. To elucidate the tured in 50% Minimum Essential Medium (MEM; Sigma) underlying mechanisms, we used a combined transcrip- and 50% Nutrient Mixture F-12, HAM (Sigma), supple- tomic-proteomic approach and utilized APP point muta- mented with 10% fetal bovine serum (FBS; PAN), 1% tions to modulate the Aβ42/Aβ40 ratio. Using a genome and non-essential amino acid solution (Sigma) and 1% L- proteome-wide approach provided us with the maximum Glutamin (Sigma), in a humidified atmosphere with 5% amount of information possible. We identified cellular CO2. We transfected 70% confluent cells with the con- retinoic acid binding protein 1 (CRABP1) as the exclusive structs described previously. transcript and protein showing strong differential expression as a consequence of an increased Aβ42/Aβ40 ratio. Preparation of cell lysates and collection of conditioned Accordingly, cells with the increased Aβ42/Aβ40 ratio media showed a reduced ability to differentiate. Remarkably, a We added 5 ml culture medium to 70% confluent cells in decreased Aβ42/Aβ40ratio did not affect CRABP1 expres- a 10 cm culture dish and conditioned media were col- sion. CRABP1 is involved in retinoic acid (RA)-induced lected after 16–48 hours. The conditioned media were differentiation [8-10] and is expected to play a crucial role centrifuged at 4°C for 1 minute at 13,000 rpm and the in neurogenesis [11]. supernatants were used for immunoprecipitation of solu- ble secreted Aβ. Cell lysates were prepared by harvesting Neurogenesis is reported to be enhanced in the hippoc- and lysing cells on ice in lysis buffer supplemented with ampi [12] of patients with AD [13] where it may produce Complete® protease inhibitor (Roche).

Page 2 of 12 (page number not for citation purposes) BMC Medicine 2008, 6:38 http://www.biomedcentral.com/1741-7015/6/38

Immunoprecipitation Proteomics: two-dimensional difference gel Conditioned media were immunoprecipitated with pro- electrophoresis tein G-Sepharose (Sigma) and the antibodies G2–10 and Briefly, dried cell pellets were solubilized in lysis buffer, G2–11. The immunoprecipitated proteins were separated centrifuged and supernatant proteins were labeled with on 12% Tris-Tricine gels. Cy3 as well as Cy5, so that each sample was labeled in a dye-switch manner. Cy2 was used as fluorophor for the Western blotting and antibodies internal standard. First dimension isoelectric focusing was Western blot analysis was performed as described else- performed on Immobilized pH-Gradient Gel Strips pH 3– where [19]. Briefly, proteins were detected with the anti- 10 Non-Linear (GE Healthcare). Second dimension body W02, specific for residues 1–10 of Aβ. sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) was performed on 12.5% isocratic 254 × Transcriptomics and data analysis 200 × 1 mm3 gels [20]. CyDye fluorescence images were Gene chip analysis was performed according to the acquired on a laser scanner (GE Healthcare) and protein Expression Analysis Technical Manual (Affymetrix) with abundance changes were analyzed with the DeCyder™ 3.0 minor modifications: Briefly, total RNA was extracted Software (GE Healthcare) [21]. For subsequent mass spec- using the Qiashredder-Kit, RNeasy Midi-columns and the trometry the proteins were stained with colloidal RNase-free DNase set (Qiagen). A total of 20 μg of RNA Coomassie Brilliant Blue [22] and protein spots were was reverse transcribed into cDNA by using oligo(dT) excised manually. primers (Proligo) and the Superscript™ Double-Stranded cDNA Synthesis Kit (Invitrogen). We in vitro transcribed Protein identification 3.3 μl of purified cDNA using the BioArray™ High Yield™ Proteins were identified as described recently [23]. Briefly, RNA Labeling Kit (Enzo Life Sciences). We fragmented 15 an automated platform [24] was used to digest the pro- μg of purified cRNA using the GeneChip® Eukaryotic teins in-gel with trypsin and to prepare the proteolytic Hybridization Control Kit (Affymetrix). We hybridized 15 peptides for matrix assisted laser desorption ionization μg of fragmented cRNA to whole genome HG-U133 A and time-of-flight mass spectrometry (MALDI-TOF-MS). For HG-U133 B oligonucleotide arrays. Chips were washed, each sample a peptide mass fingerprint (PMF) spectrum stained, scanned and the quality of the created dat-file and fragment ion spectra of up to four selected precursor images was evaluated by using MAS 5.0 and Gene Operat- ions were acquired within the same automated analysis ing Software GCOS 1.2 (Affymetrix). The quality of each loop using an Ultraflex I mass spectrometer (Bruker Dal- sample was controlled (see Additional file 1). Transcrip- tonics). Database searches were performed with the Mas- tomic data was analyzed with MAS 5.0, GCOS 1.2 (both cot Software 2.0 (Matrix Science). Only proteins Affymetrix) and Array Assist 3.3 (Stratagene). Chp-files represented by at least one peptide sequence above the were created by using the PLIER algorithm. P-values were significance threshold in combination with the presence calculated from three independent experiments using of at least four peptide masses assigned in the PMF were either a two class unpaired t-test or one-way analysis of considered as identified. variance (ANOVA). Further data analysis was performed with Excel (Microsoft). For data normalization, filtering RNA interference details and data output, see Additional files 1, 2 and 3. siRNAs were double-stranded [25,26] pre-designed, annealed Silencer™ siRNAs (Ambion). To transiently Quantitative real-time polymerase chain reaction and knock down CRABP1, 30 nM siRNA was used. After 48 selection of an endogenous control for normalization hours following transfection with siRNA, total RNA was Total RNA, was reverse transcribed into cDNA using ran- extracted from the cells and the extent of knock down was dom hexamer primers included in the High-Capacity measured by real-time polymerase chain reaction (real- cDNA Archive Kit (Applied Biosystems). This cDNA was time PCR). amplified and measured by using TaqMan® Gene expression assays (Applied Biosystems). Cycling conditions Differentiation assay were 50°C for 2 minutes, 95°C for 10 minutes, followed SH-SY5Y cells were treated with 0.1–1000 nM RA, in the by 40 cycles of 95°C for 15 seconds and 60°C for 1 absence or presence of serum for 2–10 days. Differentia- minute. Relative quantification was performed with the 2- tion was evaluated by checking the length, shape and Δ CT method. For normalization, an endogenous control number of outgrowing protrusions by phase contrast was selected out of 10 candidate controls using the Taq- microscopy at appropriate times. Phase contrast pictures Man® Human Endogenous Control Plate (Applied Biosys- were taken from living neuroblastoma cells. tems).

Page 3 of 12 (page number not for citation purposes) BMC Medicine 2008, 6:38 http://www.biomedcentral.com/1741-7015/6/38

Results We studied the transcriptomic and proteomic response to an altered Aβ42/Aβ40 ratio in human neuroblastoma cells. An increased or decreased Aβ42/Aβ40 ratio revealed differentially expressed transcripts, of which the 60 most up- regulated were used here. For the corresponding proteomic approach the 20 most up-regulated proteins were selected to validate altered protein expression. Only the overlap of transcriptomic and proteomic data was used for further analysis. To analyze altered Aβ generation in a C99Figure overexpression 1 in independent cell clones controlled manner, C99-overexpression constructs encod- C99 overexpression in independent cell clones. SH- ing the C-terminal part of APP (C99) were used [6,7]. This SY5Y cells were stably transfected with a pCEP-vector con- taining the amyloid precursor protein C-terminal fragment peptide is identical to the APP-derived C99, the ultimate C99WT, and constructs bearing the point mutations precursor for Aβ generation. C99 is processed by γ-secre- C99I45F and C99V50F. The same cell line was transfected tase in the same manner as APP-derived C99, making it an with an empty vector (negative control). Eight clones (clone ideal substrate to study γ-secretase function or its cleavage 1–3 for C99WT, clone 1–3 for C99I45F and clone 1–2 for products Aβ42 and Aβ40 without the influence of β-secre- C99V50F) with approximately similar expression levels and tase. Since, due to a point mutation, the constructs gener- C99V50F clone 3, showing stronger expression, were ated peptides only differing in a single amino acid outside selected and used for transcriptome and proteome analysis. the Aβ domain (at position 45 or 50, C99I45F and Apart from analyzing the complete set of three clones, data C99V50F, respectively) compared with the wild-type con- analysis for the transcriptomic approach was also performed struct (C99WT), they were ideal for gene expression pro- by excluding clone 3 (C99V50F) resulting in no significant difference compared to the triplicates. filing, enabling us to minimize potential technical variation influencing gene expression.

Single independent clones of the human neuroblastoma Three single independent clones each from C99WT, cell line SH-SY5Y, overexpressing C99, were selected and C99I45F and C99V50F (Figure 1) were used for transcrip- checked for Aβ42 and Aβ40 generation tomic and proteomic analyses (mock-transfected cells as SH-SY5Y cells were stably transfected with constructs cod- negative control). ing for the APP C-terminal fragment C99WT, and also with constructs bearing the point mutations C99I45F and For transcriptomics, whole genome HG-U133 A and B C99V50F and the vector only (negative control) (Figure chips were used. Replicates were prepared and hybridized 1). on different days and were derived from different independent clones. Data analysis was performed by calculat- The purpose for using these mutations was their ability to ing the mean of three independent single clones. strongly shift the Aβ42/Aβ40 ratio in either direction, as previously demonstrated in detail [6]. This was confirmed here (Figure 2).

As expected and described in detail [6,7] C99I45F and C99V50F had an opposite effect on the Aβ species generated: C99I45F is mainly processed to Aβ42, resulting in a dramatic increase of the secreted Aβ42/Aβ40 levels (relative ratio approximately 20.4 compared with the Aβ42/Aβ40 ratio in C99WT); C99V50F is mainly processed to Aβ40 (relative ratio approximately 0.3 compared with C99WT) AindependentFigureβ42 and 2 Aβ 40cell generated clones from their direct precursor C99 in [6]. Aβ42 and Aβ40 generated from their direct precursor C99 in independent cell clones. Aβ42 and Aβ40 were immunoprecipitated from conditioned media of SH-SY5Y CRABP1 was up-regulated in the mutant with an increased cells, overexpressing C99, using specific antibodies for Aβ42 Aβ42/Aβ40 ratio and Aβ40. Both Aβ species were detected by Western blot- Genome- and proteome-wide expression profiles of the ting using antibody W02. C99 is intracellularly cleaved, gen- human neuroblastoma cell line SH-SY5Y were combined erating different amounts of Aβ42 and Aβ40 in C99I45F and and compared with each other (Figure 3). C99V50F. C99I45F generates more Aβ42 than Aβ40, whereas C99V50F generates more Aβ40 than Aβ42.

Page 4 of 12 (page number not for citation purposes) BMC Medicine 2008, 6:38 http://www.biomedcentral.com/1741-7015/6/38

C99V50F. This comparison revealed an effect mediated by a changed Aβ42/Aβ40 ratio, because both mutants expressed inverse levels of Aβ42 and Aβ40 respectively (Table 2).

However, neurofilament 3 (NEF3), neurofilament light polypeptide 68 kDa (NEFL) and internexin neuronal intermediate filament protein alpha (INA) were not differentially expressed (Table 2). We regard this unaltered expression of neurofilaments as mediated by C99, since C99 is expressed in similar amounts in both mutants and hence a comparison between these two mutants results in a fold change close to 1.0 (not differentially expressed).

A comparison of SH-SY5Y cells transfected with the C99WT encoding construct versus SH-SY5Y cells trans-

OverlaprevealedAFigureβ42/Aβ 340 ofCRABP1 ratio differentially up-regulation expressed specific transcripts for an andincreased proteins fected with the empty vector (mock) provides informa- Overlap of differentially expressed transcripts and tion about the effect mediated by C99 (Table 3). proteins revealed CRABP1 up-regulation specific for an increased Aβ42/Aβ40 ratio. Comparison of C99I45F or Neurofilaments (NEF3, NEFL, INA) were down-regulated C99V50F versus C99WT revealed differentially expressed as a consequence of C99 overexpression. CRABP1 was not transcripts, of which each of the 60 most up-regulated were differentially expressed, supporting our view that C99 is used here. The 20 most up-regulated proteins each were not responsible for CRABP1 dysregulation. selected for the corresponding proteomic approach. An intersection of the transcriptomic and proteomic data was subsequently performed. Only the intersection of both Differential expression of CRABP1 was confirmed by real- approaches (four transcripts and proteins) was used for fur- time PCR ther analysis. Out of these four, only CRABP1 was up-regu- Expression of CRABP1 was measured by quantitative real- lated in C99I45F, whereas no differential expression was time PCR with cyclophilin A as an endogenous normaliza- found in C99V50F (both mutants compared to C99WT). The tion control (cyclophilin A was selected out of 10 normali- remaining three transcripts and proteins were differentially zation controls; see Additional file 1). Measurements expressed in both mutants. The proteomic approach was reflect the mean of three independent clones, measured in performed blinded by an independent laboratory. The term triplicate. The fold change for CRABP1 of mutant C99I45F 'differentially expressed' was applied when the fold change (Aβ42/Aβ40↑) compared with C99WT was 4.1 (standard exceeded a threshold of at least 1.9 either on the transcript deviation of the fold change: ± 2.3). In contrast to this, the or protein level. differential expression of CRABP1 was below our defined cut-off (<1.9) for C99V50F (Aβ42/Aβ40↓) compared with C99WT and thus was regarded as not differentially expressed. For proteomics, three clones each from C99WT, C99I45F and C99V50F were pooled, then proteins were extracted, Cells with an increased Aβ42/Aβ40 ratio up-regulated CyDye labeled and analyzed by two-dimensional differ- CRABP1, which made cells less sensitive to RA ential in-gel electrophoresis (2D-DIGE, Figure 4). CRABP1 is involved in RA metabolism and transport [27] and we found it up-regulated as a consequence of an Up-regulated proteins were identified by mass spectrome- increased Aβ42/Aβ40 ratio. This raised the question of try [24]. Only the intersection of the transcriptomic and whether cells with an increased Aβ42/Aβ40 ratio show proteomic approach was used for further analysis, thus altered responses to RA treatment. SH-SY5Y cells were sta- increasing the reliability of the data (Table 1). bly transfected with the constructs increasing or lowering the Aβ42/Aβ40 ratio (Figure 2). These cells were treated CRABP1 was the second most up-regulated protein of the with 0.1–1000 nM RA in the absence or presence of whole proteome and the second most up-regulated tran- serum. After 6 days, differentiation was evaluated by script of approximately 20,000 tested transcripts, when observing the length and number of outgrowing protru- only chip A was considered (22,283 probe sets). sions by phase contrast microscopy (Figure 5, 1A and 5, 1B). A direct comparison of both mutants revealed CRABP1 as strongly up-regulated in C99I45F compared with

Page 5 of 12 (page number not for citation purposes) BMC Medicine 2008, 6:38 http://www.biomedcentral.com/1741-7015/6/38

CRABP1Figure 4 was up-regulated specifically for an increased Aβ42/Aβ40 ratio demonstrated by 2D-DIGE CRABP1 was up-regulated specifically for an increased Aβ42/Aβ40 ratio demonstrated by 2D-DIGE. Two-dimensional polyacrylamide gel electrophoresis of CyDye-labeled proteins, extracted from SH-SY5Y cells. C99I45F and C99V50F were compared with C99WT. Differentially expressed proteins, evaluated by intensity of merged colors (Cy5, Cy3), were identified by mass spectrometry. Arrows indicate CRABP1.

Furthermore, the cell shape and number of cells were eval- CRABP1 knockdown rescued the differentiation potential uated. We selected 1 nM RA for the subsequent functional of Aβ42 overproducing human neuroblastoma cells after validation and C99I45F-transfected cells were treated with RA treatment 1 nM RA for 6 days (Figure 5, 1A). No signs of differenti- If an increased Aβ42/Aβ40 exerts the diminished differentiation were observed, irrespective of the cell confluency ation behavior via CRABP1, a CRABP1 knockdown in and duration of RA treatment (cells were checked daily by C99I45F cells should rescue this effect. We administered light microscopy for up to 10 days). In contrast to this, the 30 nM siRNA to C99I45F SH-SY5Y cells for 24 hours in cells expressing C99V50F (Figure 5, 2A) showed differen- combination with a treatment of 0.1–1000 nM (1 nM tiation at 1 nM RA treatment for 6 days: the cells were shown in Figure 5) for 2.5 to 4 days in the absence (data approximately 30–60% confluent and did not reach not shown) or presence of serum. Serum withdrawal can 100% confluency after 10 days. Cells had an average of mimic differentiation ('pseudo differentiation') and was two to four protrusions. This differentiation was observed therefore excluded from further analysis. A more than from 0.1–10 nM RA, which approximately corresponds to 50% knockdown of CRABP1 was detected by quantitative physiological plasma concentrations [28,29]. At concen- real-time PCR (p = 0.0002, n = 3). trations of 100 nM or more RA, differentiation could also be observed for the C99I45F transfected cell line. Differentiation was evaluated after 2.5 days and 4 days. Knockdown of CRABP1 in combination with 1 nM RA (Fig. 5, 2B–2D) resulted in a strong change of cell shape, whereas transfection with a nonsense sequence, combined with 1 nM RA (negative control, Figure 5, 1B–1D) did not alter the shape of the cells. The strongest differen-

Table 1: CRABP1 was up-regulated in mutant C99I45F (Aβ42/Aβ40↑) only, whereas mutant C99V50F (Aβ42/Aβ40↓) showed no differential expression of CRABP1

Name Fold change C99I45F/C99WT Fold change C99V50F/C99WT P-value

Transcriptomics Proteomics Transcriptomics Proteomics Transcriptomics Proteomics

CRABP1 2.7 2.6 1.3 -1.1 0.123 0.032 NEF3 2.6 3.1 2.3 2.7 0.038 0.01 NEFL 2.2 2.3 2.2 2.5 0.032 0.004 INA 1.8 1.8 1.6 1.9 0.056 0.002

Comparisons of both mutants with C99WT. The overlay of transcriptomics and proteomics revealed four differentially expressed transcripts and proteins respectively. Out of these four, only CRABP1 was differentially expressed in C99I45F whereas C99V50F showed no differential expression of CRABP1 (compared with C99WT). One-way analysis of variance (ANOVA) was performed for C99WT, C99I45F and C99V50F. CRABP1, cellular retinoic acid binding protein 1 (NCBI accession of the protein identified by proteomics: gi|48146151); NEF3, neurofilament 3 (gi|67678152); NEFL, neurofilament light polypeptide 68 kDa (gi|105990539); INA, internexin neuronal intermediate filament protein alpha (gi|14249342).

Page 6 of 12 (page number not for citation purposes) BMC Medicine 2008, 6:38 http://www.biomedcentral.com/1741-7015/6/38

Table 2: Direct comparison between the two mutants (C99I45F versus C99V50F) showed CRABP1 as up-regulated in C99I45F (Aβ42/ Aβ40↑) whereas neurofilaments were not differentially expressed

Name Fold change C99I45F/C99V50F P-value

Transcriptomics Proteomics Transcriptomics Proteomics

CRABP1 2.3 2.8 0.188 0.059 NEF3 1.1 -1.1 0.790 0.56 NEFL 1.1 -1.1 0.640 0.51 INA 1.1 -1.1 0.679 0.24

Direct comparison of C99I45F and C99V50F (baseline experiment, C99V50F) revealed effects mediated by an altered Aβ42/Aβ40 ratio for CRABP1 by a fold change distinctly deviating from 1.0, and an effect mediated by C99 for neurofilament 3 (NEF3), neurofilament light polypeptide 68 kDa (NEFL) and internexin neuronal intermediate filament protein alpha (INA) by a fold change close to 1.0, because C99 was approximately equally expressed in both mutants. Significance was determined performing an unpaired t-test for the direct comparison of both mutants. As to be expected, p-values were high for not differentially expressed genes [62,63]. tiation was observed at 1 nM RA. No differentiation could tor (RAR)-related orphan receptor B (RORB) was down- be observed for treatment with siRNA, but without RA or regulated 2.0-fold (p = 0.049, n = 3) in C99V50F (com- treatment with 1 nM RA, but without siRNA (data not pared with C99WT), whereas it was not differentially shown). After CRABP1 knockdown and RA-treatment the expressed in C99I45F. RAR beta (RARB) was not differen- cells were approximately 30–80% confluent (Figure 5, tially regulated in C99V50F, whereas it was up-regulated 2B–2D) and did not reach 100% confluency after 10 days. 1.4-fold (p = 0.05, n = 3) in C99I45F (compared with The extent of interconnections between cells was clearly C99WT). increased (Figure 5, 2D) compared with the negative control (Figure 5, 1D). See Additional file 1 for an enlarged Discussion picture. Human instead of murine cells were used for transcriptomics and proteomics to facilitate potential comparabil- Three further genes associated with RA metabolism were ity with patient-derived data sets. The human differentially expressed as a consequence of a changed neuroblastoma cell line SH-SY5Y has characteristics close

Aβ42/Aβ40 ratio and may have influenced the effects to primary neurons, is used to demonstrate differentiation mediated by RA processes [30-33] and is a frequently used neural cell line Three further genes may have influenced the effects medi- for microarray studies [34-37]. ated by RA. Chip analysis revealed the following differential expression: Cytochrome P450 family 26 subfamily B Association between AD and RA polypeptide 1 (CYP26B1), a RA-metabolizing enzyme Associations between AD and RA transport and metabo- [27], was found to be up-regulated 1.8-fold (p = 0.01, n = lism are known [38,39]. It was shown that disruption of 3) in C99I45F (Aβ42/Aβ40↑), whereas C99V50F (Aβ42/ the retinoid signaling pathway causes a deposition of Aβ Aβ40↓) showed no differential expression (compared with in the adult rat brain [40]. RA amounts are determined by C99WT). Direct comparison of both mutants (C99I45F/ many regulatory proteins, such as retinoid binding pro- C99V50F) revealed a 2.6-fold up-regulation for CYP26B1 teins, retinoid anabolizing and catabolizing enzymes in mutant C99I45F (p = 0.02, n = 3). Retinoic acid recep- [41]. CYP26B1 has been linked to AD and psychosis [42].

Table 3: CRABP1 was not differentially expressed in consequence of C99-overexpression in contrast to neurofilaments

Name Fold change C99WT/mock P-value

Transcriptomics Proteomics Transcriptomics Proteomics

CRABP1 1.0 1.4 0.979 0.042 NEF3 -3.4 -1.8 0.024 0.086 NEFL -3.0 -1.3 0.039 0.11 INA -1.9 -1.3 0.069 0.029

Comparison between C99WT and mock-transfected cells revealed effects mediated by C99. Neurofilament 3 (NEF3), neurofilament light polypeptide 68 kDa (NEFL) and internexin neuronal intermediate filament protein alpha (INA) were down-regulated as a consequence of C99- overexpression. CRABP1 was not differentially expressed (cut-off for differential expression at least 1.9 on the transcript or protein level, respectively).

Page 7 of 12 (page number not for citation purposes) BMC Medicine 2008, 6:38 http://www.biomedcentral.com/1741-7015/6/38

theirFigureIncreased differentiation 5 Aβ42/Aβ40 potential ratio reduced responsiveness of SH-SY5Y cells to RA and knocking down up-regulated CRABP1 rescued Increased Aβ42/Aβ40 ratio reduced responsiveness of SH-SY5Y cells to RA and the knock down of up-regulated CRABP1 rescued their differentiation potential. Phase contrast images showing living human neuroblastoma cells (SH- SY5Y), grown on collagen coated glass cover slips and treated with 1 nM retinoic acid (RA). Differentiation was evaluated by the number, shape and length of outgrowing protrusions: (1A) C99I45F (Aβ42/Aβ40↑); (2A) C99V50F (Aβ42/Aβ40↓). Differenti- ation was evaluated after RA-treatment for 6 days. Both cultures were 50% confluent when RA was added (day zero). C99I45F reached 90–100% confluency after 4–6 days without any signs of differentiation, whereas C99V50F did not exceed more than 60–70% confluency (after 6–10 days) but showed strong differentiation. C99I45F was also evaluated at 60–70% of confluency showing no signs of differentiation (data not shown), thus strong confluency of C99I45F (shown here) does not conceal puta- tive signs of differentiation. (B) C99I45F (Aβ42/Aβ40↑): 30 nM siRNA was administered to the cells for 24 hours in combination with a treatment of 1 nM RA for 2.5 days. After 2.5 days, the effects of more than 50% knockdown of CRABP1 (2B) was compared with a nonsense sequence (negative control, (1B)). (C) C99I45F: same conditions as in (B) except that RA was administered for 4 days. Differentiation was evaluated after 4 days. Knockdown of CRABP1 (2C) was compared with a nonsense sequence (negative control, (1C)). (D) C99I45F: same conditions as in (C), but with another preparation from the same experiment as in (C). (B) and (C) show preparations from different experiments. Experiments were repeated three times with con- sistent results.

One crucial mechanism whereby the availability of RA is and in cells with low CRABP1 expression RA enters the regulated is by binding to CRABP1. CRABP1 is a protein nucleus and binds to RARs [8-10]. with a molecular weight of 15.4 kDa, localized in the cyto- plasm. The gene is strongly conserved in evolution and is An association between CRABP1 and Aβ has not yet been assumed to play an important role in RA-mediated differ- established. In this study we have demonstrated that an entiation and proliferation processes. It may regulate the increased Aβ42/Aβ40 ratio resulted in CRABP1 up-regula- access of RA to the nuclear RARs. In the adult brain the tion. Furthermore, we demonstrated that up-regulated two main regions of RA signaling are the olfactory bulb CRABP1 reduced the differentiation potential of SH-SY5Y and the hippocampus [43]; both regions are predomi- cells. C99I45F-transfection of SH-SY5Y cells resulted in nantly affected in late onset Alzheimer's disease (LOAD) differentiation only if exposed to 100 nM or more of RA, [41]. CRABP1 and RA are inversely regulated [44]. but the same cell line showed already strong differentia- CRABP1 binds RA and prevents its entering the nucleus tion at 1 nM RA when CRABP1 was knocked down by

Page 8 of 12 (page number not for citation purposes) BMC Medicine 2008, 6:38 http://www.biomedcentral.com/1741-7015/6/38

more than 50%. Therefore, we estimate that a 50% knock- related proliferation and differentiation properties with down of CRABP1 makes cells more sensitive to RA by neural stem cells. We observed increased proliferation of approximately a factor of 101-102. The physiological human neuroblastoma cells in consequence of an plasma concentration of RA in humans is approximately increased Aβ42/Aβ40 ratio via CRABP1 and suggest that 10 nM and 8.4 pmol g-1 in the hippocampi of mice [45]. this influences neurogenesis by promoting proliferation. Excess of exogenous RA may over-saturate the binding However, the newly generated neurons may be prevented capacities of CRABP1 allowing the remaining RA to bind from adopting a functional phenotype, as a consequence to the RARs [46]. This provides an explanation for our of CRABP1 up-regulation restricting the quantity of RA. finding that treatment with an excess of RA (>100 nM) This view is supported by a study showing that RA induces makes no difference in the differentiation behavior detect- neurite outgrowth in SH-SY5Y cells [58]. Theoretically, it able, but differences are evident at low (physiological) seems possible that CRABP1 knock-down would release levels of RA. CRABP1 transfection of AMC-HN-7 cells the block of terminal differentiation of neurons in AD and results in an increased CYP26-mediated catabolism of RA thus improve the differentiation of neural stem cells into [27]. This decreases the RA level accessible to the nuclear a functional phenotype. RA has often been used to termi- receptors. Indeed, we found CYP26B1 to be up-regulated nally differentiate neuroblastoma cells [59,60] as well as in C99I45F, but not in C99V50F. RORB was down-regu- primary neuroblasts [61]. We observed outgrowing pro- lated in C99V50F, but not in C99I45F. Furthermore RARB trusions typical for RA-induced differentiation. Moreover, was not differentially regulated in C99V50F, but up-regu- we observed that growth stopped or slowed down in a RA lated in C99I45F. These observations might reflect a concentration depending manner, which is characteristic response of the cells to an increased RA level in C99V50F of an effective differentiation process. or a decreased RA level in C99I45F, respectively. An inverse regulation of receptors and their ligands is often Our study focused on an increased Aβ42/Aβ40 ratio, which observed [47]. is typical for AD. It does not allow us distinguish between pure Aβ42 and pure Aβ40 effects, because intracellular Linkage of the chromosomal locus 15q24 to mental processing by γ-secretase typically generates less Aβ40 retardation when more Aβ42 is generated and vice versa. However, we CRABP1 is located on the same chromosomal locus emphasize that our approach better resembles in vivo con- (15q24) as alpha polypeptide 3, 4 and 5 of the nicotinic ditions than approaches in which Aβ42 or Aβ40 is added cholinergic receptor (nAChR) and cytochrome P450, fam- from outside the cells. In our approach C99 is intracellu- ily 11, subfamily A, polypeptide 1 (cholesterol side chain larly cleaved resulting in different Aβ42/Aβ40 levels, which cleavage, CYP11A1). Association of nAChR and AD has are released into the extracellular space. This is closer to in been described previously [48]. Moreover, there has been vivo conditions than treating cells artificially with Aβ42 or found to be a linkage of the chromosomal locus 15q24 to Aβ40. mental retardation [49] and linkage of the flanking regions (15q22 and 15q26) to AD [50,51]. This linkage The generation of different Aβ42/Aβ40 ratios is inherently may be explained by the presence of alpha polypeptide 3, accompanied by the generation of other C99 cleavage 4 and 5 of the nAChR, or of CRABP1, located on the same products like the p3 peptides, the APP intracellular chromosomal locus. domains (AICDs) and further cleavage products. The C99 point mutations are expected to equally shift the p340/ Neurofilaments were inversely regulated by C99 and Aβ42, p342 and Aβ42/Aβ40 ratios, but little is known about how Aβ40 these mutations affect the AICD production. Neurons We observed down-regulation of the neurofilaments produce very little p3 from C99, and the AICD sequence NEF3, NEFL and INA as a result of C99 overexpression. starts at the ε- and not at the γ-site of APP, therefore it Interestingly, these three neurofilaments were up-regu- would be expected that the main effect of the mutations lated in response to Aβ42 and Aβ40 overproduction. This analyzed is due to altered Aβ generation. This, however, may indicate a role of NEF3, NEFL and INA in the axonal does not exclude the possibility that several C99 cleavage 'clogging' phenomenon [52-55] observed in neurons products work in concert with each other. induced by APP or its cleavage products [56]. In summary, we found that an increased Aβ42/Aβ40 ratio Sensitive balance between proliferation and up-regulated CRABP1, reducing the availability of free RA. differentiation was influenced by an altered Aβ42/Aβ40 ratio This resulted in an increased tendency towards prolifera- via CRABP1 tion accompanied by a reduced potential to differentiate. Treating neural stem cells with Aβ increases the total This effect could be rescued by knocking down CRABP1. number of neurons in a dose-dependent manner [57]. In We speculate that Aβ42 induces the initial steps in neuro- our study we used neuroblastoma cells, which share genesis by boosting neuronal precursor cell proliferation

Page 9 of 12 (page number not for citation purposes) BMC Medicine 2008, 6:38 http://www.biomedcentral.com/1741-7015/6/38

while preventing the terminal differentiation into mature neurons. This scenario may provide an explanation for Additional file 2 why in AD there is an increase in neurogenesis and at the Table 4. Data. same time an increased risk for neurodegeneration. Click here for file [http://www.biomedcentral.com/content/supplementary/1741- 7015-6-38-S2.xls] Conclusion We conclude that the differentiation potential of the Additional file 3 human neuroblastoma cell line SH-SY5Y is reduced via Table 5. Data. CRABP1 up-regulation as a consequence of an increased Click here for file [http://www.biomedcentral.com/content/supplementary/1741- Aβ42/Aβ40 ratio. 7015-6-38-S3.xls] Abbreviations 2D-DIGE: two-dimensional differential in-gel electrophoresis; Aβ: β-amyloid peptides; AD: Alzheimer's disease; AICD: APP intracellular domain; ANOVA: analysis Acknowledgements We thank Dr. Peter Prior and Cathy Ludwig for carefully reading the man- of variance; APP: amyloid precursor protein; C99: C-ter- uscript. This work was supported in part by grants from the NGFN, Deut- minal fragment of APP; CRABP1: cellular retinoic acid sche Forschungsgemeinschaft, Bundesministerium für Bildung, Forschung, binding protein 1; FBS: fetal bovine serum; INA: intern- Wissenschaft und Technologie and the European Union. exin neuronal intermediate filament protein alpha; MALDI-TOF-MS: matrix assisted laser desorption ioniza- References tion time-of-flight mass spectrometry; MEM: Minimum 1. Bertram L, Tanzi RE: Of replications and refutations: the status of Alzheimer's disease genetic research. Curr Neurol Neurosci Essential Medium; NEF3: neurofilament 3; NEFL: neuro- Rep 2001, 1:442-450. filament: light polypeptide 68 kDa; real-time PCR: real- 2. Saunders AM, Strittmatter WJ, Schmechel D, George-Hyslop PH, time polymerase chain reaction; PMF: peptide mass fin- Pericak-Vance MA, Joo SH, Rosi BL, Gusella JF, Crapper-MacLachlan DR, Alberts MJ, et al.: Association of apolipoprotein E allele gerprint; RA: all-trans retinoic acid; RAR: retinoic acid epsilon 4 with late-onset familial and sporadic Alzheimer's receptor; RARB: RAR beta; RORB: RAR-related orphan disease. Neurology 1993, 43:1467-1472. receptor B; SDS-PAGE: sodium dodecyl sulfate polyacryla- 3. Farrer LA, Cupples LA, Haines JL, Hyman B, Kukull WA, Mayeux R, Myers RH, Pericak-Vance MA, Risch N, van Duijn CM: Effects of mide gel electrophoresis. age, sex, and ethnicity on the association between apolipoprotein E genotype and Alzheimer disease. A meta-analysis. APOE and Alzheimer Disease Meta Analysis Consortium. Competing interests JAMA 1997, 278:1349-1356. The authors declare that they have no competing interests. 4. Walsh DM, Klyubin I, Fadeeva JV, Cullen WK, Anwyl R, Wolfe MS, Rowan MJ, Selkoe DJ: Naturally secreted oligomers of amyloid beta protein potently inhibit hippocampal long-term poten- Authors' contributions tiation in vivo. Nature 2002, 416:535-539. MU designed the project, wrote the paper, performed gene 5. Walsh DM, Klyubin I, Fadeeva JV, Rowan MJ, Selkoe DJ: Amyloid- beta oligomers: their production, toxicity and therapeutic expression profiling, real-time PCR, differentiation assays inhibition. Biochem Soc Trans 2002, 30:552-557. and analyzed the data. PB performed 2D-DIGE and ana- 6. Grimm HS, Beher D, Lichtenthaler SF, Shearman MS, Beyreuther K, lyzed the data. OJ performed mass spectrometry and ana- Hartmann T: gamma-Secretase cleavage site specificity differs for intracellular and secretory amyloid beta. J Biol Chem 2003, lyzed the data. MO was involved with the proteomics 278:13077-13085. project design. YK contributed to data analysis. AW made 7. Lichtenthaler SF, Multhaup G, Masters CL, Beyreuther K: A novel a technical contribution to the microarray analysis. LB substrate for analyzing Alzheimer's disease gamma-secretase. FEBS Lett 1999, 453:288-292. performed real-time PCR. KH validated the protein 8. Dencker L, Annerwall E, Busch C, Eriksson U: Localization of spe- expression. TH designed and supervised the project, and cific retinoid-binding sites and expression of cellular retinoic- acid-binding protein (CRABP) in the early mouse embryo. edited the manuscript. Development 1990, 110:343-352. 9. Dencker L, Gustafson AL, Annerwall E, Busch C, Eriksson U: Retin- Additional material oid-binding proteins in craniofacial development. J Craniofac Genet Dev Biol 1991, 11:303-314. 10. Vaessen MJ, Meijers JH, Bootsma D, Van Kessel AG: The cellular retinoic-acid-binding protein is expressed in tissues associ- Additional file 1 ated with retinoic-acid-induced malformations. Development Supplemental data 1. 1. Larger scale of Figure 5D; 2. Transcriptomic 1990, 110:371-378. data analysis;. 3. Remarks; 4. Materials and methods; 5. Quality control 11. Wilson LJ, Myat A, Sharma A, Maden M, Wingate RJ: Retinoic acid is a potential dorsalising signal in the late embryonic chick of cells, target-RNA and arrays; 6. RNA-quality assessed by using the Agi- hindbrain. BMC Dev Biol 2007, 7:138. lent 2100 Bioanalyzer. 12. Kempermann G, Chesler EJ, Lu L, Williams RW, Gage FH: Natural Click here for file variation and genetic covariance in adult hippocampal neu- [http://www.biomedcentral.com/content/supplementary/1741- rogenesis. Proc Natl Acad Sci USA 2006, 103:780-785. 7015-6-38-S1.doc] 13. Jin K, Peel AL, Mao XO, Xie L, Cottrell BA, Henshall DC, Greenberg DA: Increased hippocampal neurogenesis in Alzheimer's disease. Proc Natl Acad Sci USA 2004, 101:343-347.

Page 10 of 12 (page number not for citation purposes) BMC Medicine 2008, 6:38 http://www.biomedcentral.com/1741-7015/6/38

14. Jin K, Minami M, Lan JQ, Mao XO, Batteur S, Simon RP, Greenberg 33. Yu WH, Fraser PE: S100beta interaction with tau is promoted DA: Neurogenesis in dentate subgranular zone and rostral by zinc and inhibited by hyperphosphorylation in Alzhe- subventricular zone after focal cerebral ischemia in the rat. imer's disease. J Neurosci 2001, 21:2240-2246. Proc Natl Acad Sci USA 2001, 98:4710-4715. 34. Sonnenfeld KH, Ishii DN: Fast and slow nerve growth factor 15. Jin K, Galvan V, Xie L, Mao XO, Gorostiza OF, Bredesen DE, Green- binding sites in human neuroblastoma and rat pheochromo- berg DA: Enhanced neurogenesis in Alzheimer's disease cytoma cell lines: relationship of sites to each other and to transgenic (PDGF-APPSw, Ind) mice. Proc Natl Acad Sci USA neurite formation. J Neurosci 1985, 5:1717-1728. 2004, 101:13363-13367. 35. Kondo M, Shibata T, Kumagai T, Osawa T, Shibata N, Kobayashi M, 16. Verret L, Jankowsky JL, Xu GM, Borchelt DR, Rampon C: Alzhe- Sasaki S, Iwata M, Noguchi N, Uchida K: 15-Deoxy-Delta(12,14)- imer's-type amyloidosis in transgenic mice impairs survival prostaglandin J(2): the endogenous electrophile that induces of newborn neurons derived from adult hippocampal neuro- neuronal apoptosis. Proc Natl Acad Sci USA 2002, 99:7367-7372. genesis. J Neurosci 2007, 27:6771-6780. 36. Martinez T, Pascual A: Identification of genes differentially 17. Biedler JL, Roffler-Tarlov S, Schachner M, Freedman LS: Multiple expressed in SH-SY5Y neuroblastoma cells exposed to the neurotransmitter synthesis by human neuroblastoma cell prion peptide 106–126. Eur J Neurosci 2007, 26:51-59. lines and clones. Cancer Res 1978, 38:3751-3757. 37. Reimertz C, Kogel D, Rami A, Chittenden T, Prehn JH: Gene 18. Ross RA, Spengler BA, Biedler JL: Coordinate morphological and expression during ER stress-induced apoptosis in neurons: biochemical interconversion of human neuroblastoma cells. induction of the BH3-only protein Bbc3/PUMA and activa- J Natl Cancer Inst 1983, 71:741-747. tion of the mitochondrial apoptosis pathway. J Cell Biol 2003, 19. Ida N, Hartmann T, Pantel J, Schroder J, Zerfass R, Forstl H, Sandbrink 162:587-597. R, Masters CL, Beyreuther K: Analysis of heterogeneous A4 pep- 38. Husson M, Enderlin V, Delacourte A, Ghenimi N, Alfos S, Pallet V, tides in human cerebrospinal fluid and blood by a newly Higueret P: Retinoic acid normalizes nuclear receptor medi- developed sensitive Western blot assay. J Biol Chem 1996, ated hypo-expression of proteins involved in beta-amyloid 271:22908-22914. deposits in the cerebral cortex of vitamin A deprived rats. 20. Laemmli UK: Cleavage of structural proteins during the Neurobiol Dis 2006, 23:1-10. assembly of the head of bacteriophage T4. Nature 1970, 39. Goodman AB, Pardee AB: Evidence for defective retinoid trans- 227:680-685. port and function in late onset Alzheimer's disease. Proc Natl 21. Steinacker P, Schwarz P, Reim K, Brechlin P, Jahn O, Kratzin H, Aitken Acad Sci USA 2003, 100:2901-2905. A, Wiltfang J, Aguzzi A, Bahn E, et al.: Unchanged survival rates of 40. Corcoran JP, So PL, Maden M: Disruption of the retinoid signal- 14-3-3gamma knockout mice after inoculation with patho- ling pathway causes a deposition of amyloid beta in the adult logical prion protein. Mol Cell Biol 2005, 25:1339-1346. rat brain. Eur J Neurosci 2004, 20:896-902. 22. Neuhoff V, Arold N, Taube D, Ehrhardt W: Improved staining of 41. Goodman AB: Retinoid receptors, transporters, and metabo- proteins in polyacrylamide gels including isoelectric focusing lizers as therapeutic targets in late onset Alzheimer disease. gels with clear background at nanogram sensitivity using J Cell Physiol 2006, 209:598-603. Coomassie Brilliant Blue G-250 and R-250. Electrophoresis 42. Bacanu SA, Devlin B, Chowdari KV, DeKosky ST, Nimgaonkar VL, 1988, 9:255-262. Sweet RA: Linkage analysis of Alzheimer disease with psycho- 23. Werner HB, Kuhlmann K, Shen S, Uecker M, Schardt A, Dimova K, sis. Neurology 2002, 59:118-120. Orfaniotou F, Dhaunchak A, Brinkmann BG, Mobius W, et al.: Prote- 43. Jacobs S, Lie DC, DeCicco KL, Shi Y, DeLuca LM, Gage FH, Evans RM: olipid protein is required for transport of sirtuin 2 into CNS Retinoic acid is required early during adult neurogenesis in myelin. J Neurosci 2007, 27:7717-7730. the dentate gyrus. Proc Natl Acad Sci USA 2006, 103:3902-3907. 24. Jahn O, Hesse D, Reinelt M, Kratzin HD: Technical innovations 44. Ruberte E, Dolle P, Chambon P, Morriss-Kay G: Retinoic acid for the automated identification of gel-separated proteins by receptors and cellular retinoid binding proteins. II. Their dif- MALDI-TOF mass spectrometry. Anal Bioanal Chem 2006, ferential pattern of transcription during early morphogene- 386:92-103. sis in mouse embryos. Development 1991, 111:45-60. 25. Tuschl T: Functional genomics: RNA sets the standard. Nature 45. Kane MA, Chen N, Sparks S, Napoli JL: Quantification of endog- 2003, 421:220-221. enous retinoic acid in limited biological samples by LC/MS/ 26. Tuschl T, Zamore PD, Lehmann R, Bartel DP, Sharp PA: Targeted MS. Biochem J 2005, 388:363-369. mRNA degradation by double-stranded RNA in vitro. Genes 46. Morriss-Kay G: Retinoic acid and craniofacial development: Dev 1999, 13:3191-3197. molecules and morphogenesis. Bioessays 1993, 15:9-15. 27. Won JY, Nam EC, Yoo SJ, Kwon HJ, Um SJ, Han HS, Kim SH, Byun Y, 47. Liu KX, Kato Y, Kino I, Nakamura T, Sugiyama Y: Ligand-induced Kim SY: The effect of cellular retinoic acid binding protein-I downregulation of receptor-mediated clearance of hepato- expression on the CYP26-mediated catabolism of all-trans cyte growth factor in rats. Am J Physiol 1998, 275:E835-E842. retinoic acid and cell proliferation in head and neck squa- 48. Kawamata J, Shimohama S: Association of novel and established mous cell carcinoma. Metabolism 2004, 53:1007-1012. polymorphisms in neuronal nicotinic acetylcholine receptors 28. Ross AC: Cellular metabolism and activation of retinoids: with sporadic Alzheimer's disease. J Alzheimers Dis 2002, roles of cellular retinoid-binding proteins. Faseb J 1993, 4:71-76. 7:317-327. 49. Mitchell SJ, McHale DP, Campbell DA, Lench NJ, Mueller RF, Bundey 29. Napoli JL, Posch KP, Fiorella PD, Boerman MH: Physiological SE, Markham AF: A syndrome of severe mental retardation, occurrence, biosynthesis and metabolism of retinoic acid: spasticity, and tapetoretinal degeneration linked to chromo- evidence for roles of cellular retinol-binding protein (CRBP) some 15q24. Am J Hum Genet 1998, 62:1070-1076. and cellular retinoic acid-binding protein (CRABP) in the 50. Scott WK, Hauser ER, Schmechel DE, Welsh-Bohmer KA, Small GW, pathway of retinoic acid homeostasis. Biomed Pharmacother Roses AD, Saunders AM, Gilbert JR, Vance JM, Haines JL, et al.: 1991, 45:131-143. Ordered-subsets linkage analysis detects novel Alzheimer 30. Garzon DJ, Fahnestock M: Oligomeric amyloid decreases basal disease loci on chromosomes 2q34 and 15q22. Am J Hum Genet levels of brain-derived neurotrophic factor (BDNF) mRNA 2003, 73:1041-1051. via specific downregulation of BDNF transcripts IV and V in 51. Blacker D, Bertram L, Saunders AJ, Moscarillo TJ, Albert MS, Wiener differentiated human neuroblastoma cells. J Neurosci 2007, H, Perry RT, Collins JS, Harrell LE, Go RC, et al.: Results of a high- 27:2628-2635. resolution genome screen of 437 Alzheimer's disease fami- 31. Desbarats J, Birge RB, Mimouni-Rongy M, Weinstein DE, Palerme JS, lies. Hum Mol Genet 2003, 12:23-32. Newell MK: Fas engagement induces neurite growth through 52. Williamson TL, Cleveland DW: Slowing of axonal transport is a ERK activation and p35 upregulation. Nat Cell Biol 2003, very early event in the toxicity of ALS-linked SOD1 mutants 5:118-125. to motor neurons. Nat Neurosci 1999, 2:50-56. 32. Li L, Hung AC, Porter AG: Secretogranin II: a key AP-1-regu- 53. Sasaki S, Warita H, Abe K, Iwata M: Slow component of axonal lated protein that mediates neuronal differentiation and transport is impaired in the proximal axon of transgenic protection from nitric oxide-induced apoptosis of neuroblas- mice with a G93A mutant SOD1 gene. Acta Neuropathol (Berl) toma cells. Cell Death Differ 2008, 15:879-888. 2004, 107:452-460.

Page 11 of 12 (page number not for citation purposes) BMC Medicine 2008, 6:38 http://www.biomedcentral.com/1741-7015/6/38

54. Shah JV, Cleveland DW: Slow axonal transport: fast motors in the slow lane. Curr Opin Cell Biol 2002, 14:58-62. 55. Perez-Olle R, Jones ST, Liem RK: Phenotypic analysis of neurofilament light gene mutations linked to Charcot-Marie-Tooth disease in cell culture models. Hum Mol Genet 2004, 13:2207-2220. 56. Gunawardena S, Goldstein LS: Disruption of axonal transport and neuronal viability by amyloid precursor protein mutations in Drosophila. Neuron 2001, 32:389-401. 57. Lopez-Toledano MA, Shelanski ML: Neurogenic effect of beta- amyloid peptide in the development of neural stem cells. J Neurosci 2004, 24:5439-5444. 58. Miloso M, Villa D, Crimi M, Galbiati S, Donzelli E, Nicolini G, Tredici G: Retinoic acid-induced neuritogenesis of human neuroblastoma SH-SY5Y cells is ERK independent and PKC dependent. J Neurosci Res 2004, 75:241-252. 59. Clagett-Dame M, McNeill EM, Muley PD: Role of all-trans retinoic acid in neurite outgrowth and axonal elongation. J Neurobiol 2006, 66:739-756. 60. Jamsa A, Hasslund K, Cowburn RF, Backstrom A, Vasange M: The retinoic acid and brain-derived neurotrophic factor differentiated SH-SY5Y cell line as a model for Alzheimer's disease- like tau phosphorylation. Biochem Biophys Res Commun 2004, 319:993-1000. 61. Wang TW, Zhang H, Parent JM: Retinoic acid regulates postnatal neurogenesis in the murine subventricular zone-olfactory bulb pathway. Development 2005, 132:2721-2732. 62. Chu TM, Weir B, Wolfinger R: A systematic statistical linear modeling approach to oligonucleotide array experiments. Math Biosci 2002, 176:35-51. 63. Khalyfa A, Chlon T, Qiang H, Agarwal N, Cooper NG: Microarray reveals complement components are regulated in the serum-deprived rat retinal ganglion cell line. Mol Vis 2007, 13:293-308.

Pre-publication history The pre-publication history for this paper can be accessed here: http://www.biomedcentral.com/1741-7015/6/38/prepub

Publish with BioMed Central and every scientist can read your work free of charge "BioMed Central will be the most significant development for disseminating the results of biomedical research in our lifetime." Sir Paul Nurse, Cancer Research UK Your research papers will be: available free of charge to the entire biomedical community peer reviewed and published immediately upon acceptance cited in PubMed and archived on PubMed Central yours — you keep the copyright

Submit your manuscript here: BioMedcentral http://www.biomedcentral.com/info/publishing_adv.asp

Page 12 of 12 (page number not for citation purposes)